Security audit

Pooled

Security checks for vulnerabilities and agentic risk

Overview

This plugin appears to do what it says: it runs a peer-to-peer model room across the user's devices, with sensitive data-sharing risks clearly disclosed.

Install only if you are comfortable with devices in a Pooled room seeing enough of the conversation to influence answers and tool calls. Use rooms made of devices you trust, keep OpenClaw approvals enabled, and be especially careful with join mode because it sends the full OpenClaw context to the room host.

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

Detected: suspicious.dangerous_exec, suspicious.insecure_tls_verification

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: dist/index.js:3615
Evidence: run: (cmd, args) => execFileSync(cmd, args, { encoding: "utf8", timeout: 5e3, stdio: ["ignore", "pipe", "ignore"] }),

HTTPS certificate verification is disabled.

Warn

Code: suspicious.insecure_tls_verification
Location: dist/index.js:8924
Evidence: async embedRunBatch(ids2, basePos, verify = false) {